作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2020, Vol. 46 ›› Issue (9): 83-88,94. doi: 10.19678/j.issn.1000-3428.0055311

• 人工智能与模式识别 • 上一篇    下一篇

基于STDP规则的脉冲神经网络研究

庄祖江, 房玉, 雷建超, 刘栋博, 王海滨   

  1. 西华大学 电气与电子信息学院, 成都 610039
  • 收稿日期:2019-06-27 修回日期:2019-09-05 发布日期:2019-09-17
  • 作者简介:庄祖江(1995-),男,硕士研究生,主研方向为脉冲神经网络、图像识别;房玉,讲师、博士;雷建超,硕士研究生;刘栋博,讲师、博士;王海滨,教授、博士。
  • 基金资助:
    国家自然科学基金(61571371);西华大学大健康管理促进中心项目(DJKG2019-008)。

Research on Spiking Neural Network Based on STDP Rule

ZHUANG Zujiang, FANG Yu, LEI Jianchao, LIU Dongbo, WANG Haibin   

  1. College of Electrical Engineering and Electronic Information, Xihua University, Chengdu 610039, China
  • Received:2019-06-27 Revised:2019-09-05 Published:2019-09-17

摘要: 人类对于生物系统信息的处理主要依赖于构成复杂神经网络的数十亿个神经元,并且信息以脉冲的形式进行传输。利用STDP学习算法构建基于LIF模型的两层脉冲神经网络结构,并对分类层算法进行改进,提出一种投票竞争机制。通过多次训练后对神经元表现类别进行竞争投票,优化同等神经元数量的网络机构在图像分类问题中的性能。在MNIST数据集上进行实验验证,结果表明,该投票竞争机制准确率达到98.1%,与同等网络规模下未采用投票竞争机制的脉冲神经网络相比,准确率平均提高了约6%,而且当神经元数目较少时,在不增加训练时间情况下,可以取得与更加复杂网络结构相同的训练结果。

关键词: STDP规则, 脉冲神经网络, LIF模型, 投票竞争机制, 图像识别

Abstract: The way human beings process information mainly relies on billions of neurons that constitute a complex neural network,and the information is transmitted in the form of pulses.The STDP learning algorithm is used to construct a two-layer Spiking Neural Network(SNN) structure based on the LIF model,and a voting competition mechanism is proposed based on the improved classification layer algorithm.Through the competitive voting of neuron performance categories after multiple times of training,the performance of the network organization with the same number of neurons on the image classification problem is optimized.Results of experimental verification on the MNIST data set show that the accuracy rate of the voting competition mechanism reaches 98.1%,about 6%on average higher than that of the SNN under the same network scale without the voting competition mechanism.When the number of neurons is small,the voting competition mechanism can achieve the same training results as more complex network structures do without increasing the training time.

Key words: STDP rule, Spiking Neural Network(SNN), LIF model, voting competition mechanism, image recognition

中图分类号: